When shopping online for a Kate Spade dress, customers are prompted on the product page to “find their True Fit” below the drop-down list to a pick a size. After answering a few questions — height and weight, the name of a brand that typically fits well — True Fit returns the size that would most likely fit from Kate Spade’s offerings. It also pairs the match with a rating of how well the dress is expected to fit.

Kate Spade has integrated True Fit’s algorithm into its online store since May 2015, and Mary Beech, the brand’s evp and CMO, has said that in the time since, customer return rate has dropped and e-commerce sales have increased, both in the double digits, although the company does not break out specific figures. Beech said that the goal of the integration was to ensure customer confidence online. True Fit’s other retail customers include Nordstrom, Adidas, Topshop and Macy’s.

“Body types are unique, and that’s being addressed more and more in retail,” said Jane Hali, a retail research analyst. “Fit has been difficult forever, but it’s really important for retail today. E-commerce companies see money going down the drain.”

According to Hali’s research firm, Jane Hali & Associates, 2015 saw a total of $250.5 billion of lost sales due to returns across all retail in the U.S. To solve for fit online, companies are looking for ways to increase customer confidence and drive down return rates. There are algorithm integrations, like True Fit, business models that optimize for the uncertainty of an online apparel purchase, and social media hacks being thrown at the wall in order to solve for fit, a point that’s become something to compete on among online retailers.

“There’s likely not going to be one fit solution that becomes the standard across the retail industry,” said Harrison Lewin, associate analyst at digital firm L2. “It’s going to be a strength — the retailers who approach this well will be the ones who excel.”

How data helps
True Fit mines data from millions of product specs, what CEO and co-founder Romney Evans calls the “blueprint” of apparel and footwear. From there, the algorithm combines the specs with style attributes, which is the product information that’s more anecdotal and less likely to be picked up with hard data, like a stretchy material. Then it compares those attributes with human body type, and makes an educated guess as to whether the two are a match, or in other words, a good fit. The service currently has 22 million registered users — meaning, those who have taken the True Fit quiz and saved the results to a profile — and it recently raised a $25 million round of funding.
Screen Shot 2016-08-18 at 11.33.12 AM

True Fit results on Kate Spade’s website

“We’re constantly collecting data from people and garments,” said Evans. “And then we understand a pattern. But the important part is that it only asks for about 30 seconds of the shoppers’ time.”

However, those 30 seconds might feel like a long time, even as customers are on a retailer’s product page, pondering a purchase. It’s not a guarantee that they’ll opt in to True Fit at all.

“When people find a True Fit match, that’s a big confidence booster,” said Lewin. “But it’s easy to ignore that True Fit prompt on the product page, and not take that second step. Plus, retailers have to make the decision to invite other brand names to their sites. That makes them hesitant.”

When customers take the True Fit quiz, they’re asked which brand they know fits them. That can remind them that there’s other options they’re more certain about elsewhere. Instead, retailers are likely to work with a third-party vendor to round out their own data around customer behavior, like returns and successful purchases.

The power of user-generated content
If a recommendation from a service like True Fit feels to opaque to customers, user-generated content is a visual review for an item that catches customer’s attention.

“People want to see what something looks like on,” said Lewin. “The biggest piece of this is user-generated content. It’s incredibly powerful in driving people to click purchase.”

Of course, it has to be well done. Lewin said it’s not enough to have the user content on the homepage or an in isolated spot. It has to be alongside reviews on the product page as customers are making that decision. Retailers like Rent the Runway use user-uploaded content accompanied next to in-depth reviews. The company also took it a step further this summer when it launched “virtual try ons” on Snapchat, asking followers on the platform to share the dress they wanted to rent, as well as their height and size, before recruiting a Rent the Runway employee to model it and review it in a video.

“It’s really important to guarantee a great fit,” said Rent the Runway svp of merchandising Sarah Tam. “And what people respond to most are reviews from others with similar body types.”

Tam said that the user-generated photos and reviews, called Rent the Runway On You, work best when the photos look professionally done and can blend in with product shots.

Screen Shot 2016-08-15 at 10.45.24 AM

Rent the Runway On You reviews

“People like to get inspiration from places like Instagram, so if they can use that at an actual point of purchase, it’s a visual affirmation,” said Lewin. “It’s a well-placed ‘I need to buy this’ moment.”

The importance of post-purchase
While data mining and user-generated reviews are important in driving the decision to purchase, they only account for half the journey.

“Every approach to perfecting fit is valuable, and the more data a user can take into consideration while they’re evaluating a product is critical,” said Rachel Bogan, partner at the agency Work & Co. “But we find that that’s only half the story. Companies that help customers feel comfortable by promoting proper return policies are the ones who win.”

Bogan continued that offers like free shipping and free returns, in store returns, and reserve online and try in store are becoming “table stakes” for retailers to compete. While it’s important to show a customer as much information about a product as possible, customers still need to touch, feel and try on themselves, and how companies behave once that’s happened makes a big difference.

“It’s having that end-to-end cross-channel solution that gives the customer confidence,” said Bogan.

She added that the companies who rise to the top will be the ones who figure out the perfect balance of data, customer service and user experience. While there’s no “silver-bullet solution,” she said, a safe bet for retailers is to model their practices after Amazon Prime, which has mastered the easy mechanics of online shopping, like fast and free shipping and returns.

“Amazon Prime is where to look for basic practices that players will be forced to adopt to remain competitive,” said Bogan.